Autoantibodies against the thyroid gland with thyroid peroxidase antibody (TPO‐Ab) and thyroglobulin antibody (Tg‐Ab) as the most common can often be demonstrated in serum. The effect of public iodization programmes on antibody prevalence is uncertain. Aim To measure the concentrations of thyroid autoantibodies in the Danish population before and after mandatory iodization of salt. Methods Two identical cross‐sectional population studies were performed before (Cohort 1 (C1), year 1997–1998, n = 4649, median urinary iodine 61 μg/l) and 4–5 years after (Cohort 2 (C2), year 2004–2005, n = 3570, median urinary iodine 101 μg/l) mandatory iodine fortification of salt was implemented in Denmark. Blood tests were analysed for TPO‐Ab and Tg‐Ab using sensitive assays. Results Antibodies were more frequent in C2 than in C1: TPO‐Ab > 30 U/ml, C1 vs C2: 14·3 vs 23·8% (P <0·001) and Tg‐Ab > 20 U/ml, C1 vs C2: 13·7 vs 19·9% (P <0·001). The C2 vs C1 effect was confirmed in multivariate regression models (C1 reference): TPO‐Ab: OR (95% CI): 1·80 (1·59–2·04) and Tg‐Ab: 1·49 (1·31–1·69). The increase in the frequency of thyroid antibodies was most pronounced in young women and especially observed at low concentrations of antibodies. Conclusion The prevalence of both TPO‐Ab and Tg‐Ab was higher 4–5 years after a cautious iodine fortification of salt was introduced in Denmark. The increase was most pronounced in young women and in the low concentrations of antibody. Further studies are needed to evaluate the long‐term effects of increased iodine intake on thyroid autoimmunity in the population.